Azoalkanes of the 2,3-diazabicyclo[2.2.2]-oct-2-ene type have been introduced as probes for antioxidants in homogeneous solution as well as in liposomes and micelles. The bimolecular fluorescence quenching of the bridgehead dichloro-substituted 1,4-dichloro-2,3-diazabicyclo[2.2.2]-oct-2-ene (3) was compared with that of the parent compound 2,3-diazabicyclo[2.2.2]-oct-2-ene (1) and the bridgehead-dialkylated compound 4-methyl-1-isopropyl-2,3-diazabicyclo[2.2.2]-oct-2-ene (2). Compound 3 showed a more efficient fluorescence quenching in C-H containing solvents (e.g., in n-hexane: 30 ns for 3 versus 340 ns for 1 and 770 ns for 2), but a less efficient quenching in aqueous solution (e.g., in deaerated H(2)O: 485 ns for 3 versus 420 ns for 1 and 340 ns for 2), and also by molecular oxygen (k(q)/10(9) M(-1) s(-1) = 0.32 for versus 2.5 for 1 and 1.9 for 2). Towards low-molecular weight antioxidants, compound 3 showed a significantly higher reactivity (e.g., for reduced glutathione: k(q)/10(9) M(-1) s(-1) = 1.8 for 3 versus 0.82 for 1 and 0.39 for 2), at the expense of a lower differentiation between the investigated antioxidants (lower selectivity). The increased reactivity of 3 and lower, as well as qualitatively different, selectivity is attributed to a combination of factors, most importantly the slightly increased excitation energy of 3 and its lower excited-state nucleophilicity. The latter was independently corroborated, besides its longer fluorescence lifetime in aqueous solution, through the trends in quenching rate constants of the azoalkanes 1-3 towards electron-deficient versus electron-rich lactone antioxidants of the benzofuranone type. While common inorganic buffer constituents caused no fluorescence quenching, significant quenching was observed, as a curiosity, for hydrogencarbonate (k(q)/10(6) M(-1) s(-1) = 1.7 for 3 versus 2.4 for 1 and 0.45 for 2), with a fully manifested kinetic deuterium isotope effect (k(q)(H(2)O)/k(q)(D(2)O) = 12) for 3.